The Abu Dhabi Autonomous Racing League concluded its second season with a six-car autonomous Grand Final at Yas Marina Circuit, marking the largest autonomous race event to date. Germany’s TUM retained the championship title ahead of TII Racing (UAE) and PoliMOVE (Italy) in a 20-lap competition that featured AI-controlled vehicles reaching speeds exceeding 250 km/h. The event included eleven international teams competing for a $2.25 million prize pool and a human-versus-AI demonstration with former Formula 1 driver Daniil Kvyat.
Highlights
- TUM won the Grand Final after leader Unimore collided with a lapped car at mid-race, while Unimore recorded the fastest lap of the competition
- Daniil Kvyat’s best lap time of 57.57 seconds exceeded TUM’s autonomous car HAILEY by 1.58 seconds, reducing the performance gap from 10 seconds recorded 18 months ago
- Six teams qualified for the Grand Final following intensive testing that produced AI lap times surpassing human benchmark performance
- The event served as the conclusion to Abu Dhabi Autonomous Week, a six-day program combining autonomous technology summits and exhibitions
Race Results and Technical Performance
TUM started from pole position but faced immediate pressure from Unimore, which overtook for the lead at Turn 6 during the second lap. The two teams maintained separation of less than one second throughout the next ten laps, operating at speeds above 250 km/h.
At the race’s midpoint, both leaders encountered slower traffic. Unimore attempted to overtake sixth-place Constructor mid-corner, resulting in contact that sent both vehicles off track. TUM inherited the lead and maintained position through the remaining laps.
Prof. Markus Lienkamp, Team Principal at TUM, stated: “We expected a tight battle with Unimore behind us from the beginning. They have been getting faster and faster in every test session, especially on cold tyres. We expected they’d overtake us on lap two and then we would get closer as our tyres warmed up.”
Human vs AI Performance Gap
Former Formula 1 driver Daniil Kvyat competed against TUM’s autonomous vehicle HAILEY in a ten-lap pursuit challenge beginning with a 10-second gap. Kvyat recorded a best lap of 57.57 seconds compared to HAILEY’s 59.15 seconds.
The 1.58-second difference represents significant progress from previous testing:
- Initial development phase: Minutes separated human and AI performance
- 2024 Japan exhibition: 10-second gap
- Current performance: 1.58-second gap
Kvyat commented: “Looking back to when A2RL development first began a couple of years ago, with perhaps minutes between a human driver and the AI car, down to 10 seconds in our first showcase last year, and now we are seeing performance within fractions of a second – the technology progress is staggering.”
Season 2 Technical Development
Pre-race qualification sessions demonstrated AI systems closing and surpassing human benchmark lap times. Teams progressed from trailing human drivers by minutes to achieving faster lap times through iterative testing.
Marko Bertogna, Team Principal of Unimore Racing, stated: “I was very, very happy with the performance that we showed, our speed, the overtake, all showed the professional level that we have reached. Even a human racing driver would not have been able to avoid the collision we had, so this is just the nature of high-performance racing.”
Competition Structure
Six teams qualified for the Grand Final in pole position order:
- TUM (Germany)
- Unimore (Italy)
- Kinetiz (UAE)
- TII Racing (UAE)
- PoliMOVE (Italy)
- Constructor (Germany)
Eleven teams participated in total across qualifying rounds and the final event.
Industry Context and Development Platform
H.E. Faisal Al Bannai, Adviser to the UAE President and Secretary General of the Advanced Technology Research Council, stated: “A2RL shows what happens when bold ambition meets scientific discipline. It is more than a race – it is a testbed accelerating the future of autonomous systems while building public trust in the technologies that will soon move through our cities, skies and industries.”
Stephane Timpano, CEO of ASPIRE, commented: “In just 18 months, teams have gone from standing still to delivering human-beating lap times and mastering complex overtakes – progress that can take years. By combining virtual SIM Sprints with extensive real-world testing, they’ve unlocked capabilities that will influence far more than racing.”
Educational Components
A parallel STEM Competition involved 140 students from across the UAE operating DeepRacer 1/18th-scale autonomous vehicles. UAE University won the University League, while SABIS – Ras Al Khaimah claimed the High Schools title.
The Grand Final concluded Abu Dhabi Autonomous Week, a six-day event combining the Abu Dhabi Autonomous Summit, DRIFTx Exhibition, and RoboCup Asia-Pacific. Over 8,000 spectators attended the race at Yas Marina Circuit’s North Grandstand.
His Highness Sheikh Zayed bin Mohamed bin Zayed Al Nahyan, His Highness Sheikh Mohammed bin Sultan bin Khalifa Al Nahyan, and His Excellency Faisal Al Bannai presented trophies to the winners.
Sign up for our weekly emails to stay up to date on the latest news!
